International Journal of Current Engineering and Technology

E-ISSN 2277 – 4106, P-ISSN 2347 - 5161

®
©2014 INPRESSCO , All Rights Reserved
Available at http://inpressco.com/category/ijcet

Research Article
Performance Evaluation of Cluster Head Selection using ANFIS
Harpreet KaurȦ* and Ameeta SeehraȦ
Ȧ
Department of Electronics and communication Engineering, GNDEC , Ludhiana,India

Accepted 05 Nov 2014, Available online 01 Dec 2014, Vol.4, No.6 (Dec 2014)

Abstract

Wireless sensor network are emerging in various fields like environmental monitoring, mining, surveillance system,
medical monitoring. Wireless sensor networks consist of thousands of tiny nodes having the capability of sensing,
computation and wireless Communications. Leach protocol is used as low energy consumption routing protocol.
Communication between clustered data and Base station is done via Cluster head only so cluster head selection is one of
challenging issue in wireless communication. In this paper we have introduced proposed approach for cluster head
selection using neurofuzzy algorithm. Comparative analysis of proposed leach is done with leach, leach c, leach cc
.Simulation results conclude that our proposed leach can improve life time of network and save more energy as compare
to other leach protocols.

Keywords Wireless sensor network, Leach, Leach C, Leach CC, Cluster Head Selection, Network life time.

1. Introduction Ando et al ,2010). However the cluster-head selection in

LEACH protocol is lack of balancing the whole network
1
Wireless Sensor Networks (WSNs) are usually self- energy consumption, with the result that low energy nodes
organized wireless ad hoc networks comprising of a large run out of energy prematurely and decline the network
number of resource constrained sensor nodes. One of the life. This paper analyses the effectiveness of LEACH
most important tasks of these sensor nodes is systematic protocol in cluster-head selection, and proposes an
collection of data and transmits gathered data to a distant improved energy balanced clustering algorithm.
base station (BS). With its wide application in military
reconnaissance, medical aid, logistics management, 2. Introduction to Leach Protocol
environmental monitoring, agriculture and other
commercial areas, WSN has become the furthermost W. R Heinzelman proposed LEACH protocol, which
technology in the field of communication and computer focused around group structure and progressive
research (J. Anand et al,2012). Wireless Sensor Network technology. With respect to the customary protocols,
(F.Alotaibi et al, 2007) generates a large amount of data LEACH could spare a more noteworthy level of vitality.
that has to be aggregated at various levels. A For most sensor nodes, the short-separate correspondence
multidimensional aggregation approach is considered for spared more vitality, in LEACH, the more correspondence
exhibiting the node parameters for each network. was constrained in clusters, just a couple Next Node spoke
Bandwidth, memory, signal strength, time, battery power with base stations long separation. It utilized versatile
etc. have been utilized to examine the performance of a innovation and Next node pivot technology, LEACH was
sensor network. more proficient than the first class system structure; the
Hence network life- time becomes an important entire WSN was more adjusted on burden appropriation,
parameter for efficient design of data gathering schemes and could broaden the WSN lifetime extraordinarily.
for sensor networks. Energy management is a key issue in Moreover, each one cluster could compute by regional
the deployment of sensor networks, because sensors only standards and uproot repetitive information, lessen the
reply on the battery for the power which cannot be corresponding trouble of Next Node. As the vitality
recharged or replaced. The available energy is considered utilization of the count was substantially short of what the
a major factor when designing protocols in a sensor vitality utilization of correspondence, so LEACH could
network. The lifetime of the battery operated nodes may spare vitality incredibly. In any case there were still issues
be extended in some way by adopting appropriate energy in LEACH: Firstly, the node utilized force control when
conservation mechanism for efficient communication. sending information, the vitality utilization of node was
Low-Energy Adaptive Clustering Hierarchy (LEACH) is a not same. At the point when the Next Node decision, it
classical clustering routing in wireless sensor networks (H. was viewed as that the quantity of Next Node previously,
without the vitality distinction between the remained
*Corresponding author: Harpreet Kaur nodes, prompting uneven dispersion of vitality utilization.
3871 | International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec 2014)
Harpreet Kaur et al Performance Evaluation of Cluster Head Selection using ANFIS

Also, LEACH chose Next Node haphazardly, the quantity As shown in above figure 1. We have used two inputs for
of Next Node shut to the ideal worth couldn't be ensured. train anfis i.e. input 1 and input 2 and there is only one
1 Operation: LEACH operation is broken into rounds, output. Fig 2 shows the membership function of input 1
with each round having a set-up stage and a steady state i.e. energy ranges between 0 -2 joules. Figure shows the
stage. membership function for input 2 i.e. Distance to Base
Setup stage: Each node chooses whether or not to be a Station .Ranges between 0 -175 k.m.
cluster head focused around its remaining vitality and a
universally known sought rate of Next Nodes. Every node 3.2 ANFIS Training and Error Values
choosing itself as a group head shows a commercial
message proclaiming its expectation. Non-cluster head We have loaded training and testing data set for anfis
node get perhaps a few ads and pick one group to join modeling .Training data is shown by ‘o’ and FIS output is
focused around the sign quality of the commercial got shown by ‘*’ As shown in fig 5.7 ,After 40 epochs average
from the comparing cluster head. testing error for training data is observed is 0.07503.And
Steady state stage: Each cluster head holds up to get after testing data set average testing error is found is
information from all nodes in its group and afterward 0.327. As shown in fig 5.8 ,checking data is shown by
sends the accumulated or layered come about over to a BS. ‘blue dot ’ and Fis output is shown by red ‘*’
The cluster head selection in each cluster during the setup
phase ensures the data transmission in steady state phase
when there is need to transmit data, first they will send
data to Ch. during idle slot canned in TDMA schedule
table. Cluster heads should maintain communication status
at all times and it will help to receive data from different
sensor nodes. After receiving all the data sent by their
members, CHs will aggregate them firstly and then send
them to BS. When some sensor nodes may detect similar
data them data reduce unwanted bandwidth cost, energy
consumption is less. To overcome the problem of Cluster
head dies early, a new round starts and new clusters Figure 4 Training data set output
formation will be done in the sensor network.

3. Simulation Results

3.1 Overview of Anfis Structure

Figure 5 Testing data set output

Figure 1 Structure of ANFIS
As we know for ANFIS, we need data sets for training.
Data sets are the combination of inputs and outputs. When
the training process initiated epochs are set ,then after
every epoch error is minimized .We have used 40 epochs
for error minimization. The error minimization is shown in
the Figure 5.7. After 40 epochs the error is minimized to
0.639981. Figure 5.9 Shows the testing error after
training, testing data set. After 40 epochs testing error is
minimized to 0.07503.
Figure 2 Membership function of input 1

Figure 3 Membership function of input 2 Figure 6 Minimum training error

3872 | International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec 2014)
Harpreet Kaur et al Performance Evaluation of Cluster Head Selection using ANFIS

Table 1 Minimum training values

S.No. Error Value

1 Training error 0.639981

2 Testing error 0.349269

Figure 10 Surface diagram

The linguistic variable are ‘LOW’,‘MEDIUM’, ‘HIGH’

and ‘NEAR’, ‘MEDIUM’, ‘FAR’ respectively. Layer 3
Figure 7 Minimum testing error represent the number of rules generated. We have used 9
rules which are shown by blue spot. Layer 4 represents
3.3 FIS Rules and Simulation Parameters an output membership function after training the input
membership functions. Layer 5 shows the output. Final
We have made 32 =9 anfis rules .Fig 8 shows the 9 anfis output is shown only by a single node. This layer
rules .Fig 9 shows the chance to become clusterhead represents the final output using black dot.
inacordance with the variation of input 1 and input 2 .Fig
10 shows the surface diagram of input 1 and input 2 and
chance output 1.
Figure 11 shows the shows the generated ANFIS
structure using MATLAB. The two inputs that we
considered are the Energy and Distance to Base Station are
shown in layer one. Both inputs are represented by black
dots. Layer 2 represents the input memberships of energy
and distance to the Base Station.

Figure 11 ANFIS Structure

3.4 Simulaion Parameters

Table 2 Simulation Parameters

S.No Parameters Value

Figure 8 Rule editor window 1 No of Nodes 100
2 Network Size (100 x 100) m2
3 Initial Energy(Eo) 2J
4 Transmission Energy (ETX) 50nJ/bit
5 For Energy Reception (ERX) 50nJ/bit
6 Free Space Energy Loss (Efs) 10 pJ / bit /m2
7 Multi path Energy Loss (Emp) 0.0013pJ / bit
/m4
8 Maximum no. of Rounds (rmax) 700
9 Location of BS(Sink) (50,175)
10 Adaptive Energy(EDA) 5n J/bit
Figure 9 Rule viewer 11 Packet Length 4000 bits

3873 | International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec 2014)
Harpreet Kaur et al Performance Evaluation of Cluster Head Selection using ANFIS

4. Network Model
Energy consumption in WSN
100
The network modelmake available the operating
90
environment that consists of N nodes and one base station. Proposed Leach
LeachC
Nodes are randomly deployed in an area of (100 x 100) 80
LeachCC
m2. The base station is fixed at a far distance from the 70 Leach

Total Energy(joule)
sensor nodes. The sensor nodes periodically sense the 60
environment and send data to base station. In this work,
50
our model sensor network has following properties:
40
1. The Base station is immovable at far away from the
sensor nodes 30

2. The sensor nodes are homogeneous and energy 20

constrained with even energy. 10
3. The energy rate for transmitting a data packet depends
0
on the distance of transmission. 0 100 200 300 400 500 600 700
No of Rounds(seconds)
4. The communication channel is symmetric.

Alive Node in WSN Figure 14 Energy consumption of Proposed leach ,leach

100
C, Leach CC,leach.
Proposed Leach
90
LeachC
80 LeachCC Conclusion
Leach
70
This paper presents the improvement in LEACH CC and
No of Alive Nodes

60
Comparison analysis of Leach ,leach c,LEACH CC.In
50 LEACH CC cluster head is selected on bases of residual
40 energy only.In our proposed approach we have used
30 residual energy as well distance to base station for cluster
20 head selection .Hence obtained results shows the
10 significant improvement in longevity of network life and
0
packets transmitted to base station. It shows that
0 100 200 300 400 500 600 700
No of Rounds(seconds)
communication is better in proposed Leach . The proposed
Leach has momentous improvement in the network life
Figure 12 Comparative analysis Alive Nodes of Leach , span and energy saving . The comparative analysis of
Leach C,Leach CC and Proposed Leach proposed Leach is done with leach, leach c,leach cc.
As we can be seen from the above fig 12 that life time of References
proposed leach is better as compare to Leach ,Leach c and
Leach CC Protocols.Proposed leach improves the network Alotaibi,F., Abdennour,A and Ali ,A.,(2007) , A robust prediction model
life time by 84%.From the above figure 13 proposed leach using ANFIS based on recent TETRA outdoor RF measurements
transmit more data packets to base station as compare to conducted in Riyadh city, International Journal of Electronics and
Communications (AEU) ,pp.1-9
other leach protocols. So communication over the network Anand,J., Tamilsevi J. and Janakiraman S.(2012) Analyzing the
can be performed for more time. Performance of Diverse LEACH Algorithms for Wireless Sensor
Networks, In the proceedings of Int. J.Advanced Networking and
x 10
5
Total Packets Received at BS in WSN Applications (IJANA) Vol.04 ,pp. 1610-1615 .
3
Ando, H., Barolli, L., Durresi, A.,Xhaf, F.,and Koyam, A.(2010) , An
Intelligent fuzzy- based Cluster Head Selection System for WSNs and
Proposed Leach
2.5
LeachC Its Performance Evaluation for D3N Parameter, International
LeachCC Conference on Broadband, Wireless Computing, communication and
2 Leach Applications (BWCC 2010), pp.648-653. Nov4-6
No of Packets

Azadeh,A., Saberi,M., Nadimi,V., Iman,M. and Behrooznia ,A.(2010) ,

1.5 An Integrated Intelligent Neuro-Fuzzy Algorithm for Long-Term
Electricity Con-sumption: Cases of Selected EU Countries, Int.J of
1 applied sciences Acta Polytechnica Hungarica, Vol. 7, No. 4, pp. 71-
90.
0.5
Azar, A.T., (2009) Adaptive Neuro-Fuzzy System For Hemodialysis
Treatment Process, Ph.D. Dissertation, Dept. Sys. & Biomed.
Eng.,Cairo Univ., Egypt.
0
0 100 200 300 400 500 600 700 Azar, A.T. Kandil, A.H., Wahba, K. And Massoud, W. (2008), Neuro-
No of Rounds(seconds)
Fuzzy System For Post-Dialysis Urea Rebound Prediction,
Proceedings Of IEEE 4 Cairo International Biomedical Engineering
Figure 13 Comparative analysis of Total packets received Conference (Cibec'08), Dec. pp 18-20.
at BS ( Proposed ,leach c, leach cc ,leach ) Azhar , S. and . Rahman ,S.(2007) ,Application of Artificial Neural
Network in Fault Detection Study of Batch Esterification Proces,
Figure 14 shows that Proposed leach consumes less energy International Journal of Engineering & Technology IJET-IJENS Vol.
10 No: 03
as compare to other leach protocols. So Proposed leach Bajaber F.and Awan,(2011) , Adaptive decentralized re-clustering
can save more energy and prolongs the network life time protocol for wireless sensor Networks,In Proceeding of Journal of
effectively. Computer and System Sciences, Vol 77,pp 282-292.
3874 | International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec 2014)
Harpreet Kaur et al Performance Evaluation of Cluster Head Selection using ANFIS

Chakraborty,J.(2010), Anfis Based Opportunistic Power Control For Najumnissa ,D. and Rangaswamy,T.R. (2012) , Detection and
Cognitive Radio In Spectrum Sharing, Master’s Thesis, Dept.of Classification of Epileptic Seizures using Wavelet feature extraction
applied signal processing, Blekinge Institute of Technology, Sweden. and Adaptive Neuro-Fuzzy Inference System, International Journal
Erkollar,A. and Goztepe ,K.(2013), A Study on Innovation Performance Of Computational Engineering Research ,Vol 2,No 3,pp.756-761.
forecasting in Advanced Military Education Using Neuro-Fuzzy Neshat,M., Adeli ,A., Masoumi ,A. and Sargolzae M.,(2011), A
Networks, International Journal of Science and Advanced Comparative Study on ANFIS and Fuzzy Expert System Models for
Technology,Vol .3, No 4. Concrete Mix Design, International Journal of Computer Science ,
Godbole ,V.(2012) , FCA - An Approach On LEACH Protocol Of Vol. 8, No. 2, pp. 196-210 .
Wireless Sensor Networks Using Fuzzy Logic, International Journal Padmanabhan, k. and Kamalakkannan, P.(2011) , Energy Efficient
of Computer Communications and Networks (IJCCN),Vol 2,No 3 Adaptive Protocol for Clustered Wireless Sensor Networks, In
Gogoi, R.and Sarma, K.K. (2012), ANFIS based Information Extraction International Journal of Computer Science issue (IJCSI),Vol 8, Issue
using K-means Clustering for Application in Satellite Images, 5, No 1, September 2011, pp296-301.
International Journal of Computer Applications,Vol.50,No 7,pp 13-
18. Park ,J. and Sahni ,S. (2006),An Online Heuristic For Maximum Lifetime
Karimpour J. and Abad J.,(2010), A New Routing Algorithm for Routing In Wireless Sensor Networks, IEEE Trans. Computer., Vol.
Wireless Sensor Networks,International Conference on Computer 55, No. 8, pp. 1048–1056.
Applications and Industrial Electronics (ICCAIE 2010),pp.328-333. Perrig, A., Stankovic, J. and Wagner, D. (2004) ,Security In Wireless
Kaur, H.and Seehra, A.(2014), Performance Evaluation Of Energy Sensor Network, Magazine on communications of the ACM , Vol.
Efficient Clustering Protocol For cluster Head Selection In Wireless 47, No .6, pp.53-57
Sensor Network, International Journal of Peer to Peer Networks Ramesh, K. and Somasundaram, K. (2012) ,Improved Fair-Zone
(IJP2P) Vol.5, No.3, pp.1-5. technique using Mobility Prediction in WSN, International Journal Of
K.N., V .and B. P., V. (2012),Hidden Markov Model with Computational Advanced Smart Sensor Network Systems (IJASSN ), Vol. 2, No.2.
Intelligence for Dynamic Clustering in Wireless Sensor Networks, Rashid , T.A. and Ahmed ,H.A.,(2012), Building an Adaptive Neural
International Journal of Computer Science and Information Fuzzy Inference System for Revised General Test/Gross Point
Technologies, (IJCSIT), Vol. 3 No.2 ,pp.4920-4926. Average Problem, International Journal of Emerging Technologies in
Kulkarni, R.V, Forster, A. and Venayagamoorth, G.K Computational and Applied Sciences (IJETCAS) ,Vol. 1,No 3,pp.56-
(2011),Computational intelligence in wireless sensor networks: A 65.
survey, IEEE Commun. Surveys Tutorials, vol. 13, no. 1, pp. 68–96. Ray S., Demirkol I and Heinzelman W. (2010) ,Analysis and
Kumar D.,Aseri T and Patel R.,(2011) ,Energy Efficient Clustering and optimization of energy efficiency through data advertisements for
Data Aggregation Protocol for Heterogeneous Wireless Sensor wireless sensor networks- ADV-MAC, Proc. of the 5th International
Networks,In Proceeding of Int. J. of Computers, Communications & Conference. on Mobile Ad-hoc and Sensor Networks, MSN
Control,Vol 6,No 1,pp.113-124. ’09.pp.876-892.
Kumar,R. and Kumar, M. (2012), LEACH: Features, Current Shujiang, L., Kuan,D. and Lixin, K. (2011), Improvement and Simulation
Developments, Issues and Analysis, International Journal of of Clustering Routing Algorithm in Wireless Sensor Network, 6th
Computer Science and Communication Engineering, Vol.1,1 ,pp.1-5. IEEE Conference on Industrial Electronics and Applications,pp.1068-
Li, B. and Zhang, X. (2012) , Research and Improvement of LEACH 1073.
Protocol for Wireless Sensor Network, 2012 International Shukla ,K., (2013), Research On Energy Efficient Routing Protocol
Conference on Information Engineering, Vol.25. LEACH For Wireless Sensor Networks, International Journal of
Liu, Y., Zhong, S., You L., Lv ,B. and Du, L.(2012) , A Low Energy Engineering Research & Technology (IJERT),Vol.2,pp.1-5.
Uneven Cluster Protocol Design for Wireless Sensor Network ,In the Singh,D., Choudary, N. and Pandey, S. (2012) , Performance Analysis of
proceedings of Internatinal Journal of Communications Network Energy Efficient Clustering Algorithms for Wireless Sensor
and System Sciences, Vol. 5,pp .86-89. Network, Special Issue of International journal of Computer
Madheswaran, M. and Shanmugasundaram, R.N.,(2013) , Enhancements Applications on Wireless Communication and Mobile Networks,
Of Leach Algorithm For Wireless Networks,In the proceedings of Vol.12,pp.56-59.
Journal On communication Technology, Vol. 04, pp.821-827. Tandon, R. (2012), Determination of optimal no of clusters in wireless
Mahdavi, Z., Khademi, M., Khosravighasemabadi, H.(2012), Estimated sensor networks , International Journal of Computer Networks and
Necessary Time for Receiving the Information From Cluster Head to Communications IJCNC),Vol.4,No. 4.
BS by Neuro-Fuzzy, International Journal of Engineering Research Tao, L.,Xin, Z. and Luqiao, Z.(2010) ,An Improvement for LEACH
and applications(IJERA),Vol.2,No.2,pp 1024-1026. Algorithm in Wireless Sensor Network, 5th IEEE Conference on
Mahizh ,P., Muthuselvi and Kavitha,V.(2013), A Maximum Residual Industrial Electronics and Applications , pp. 1811-1814.
Energy for WSN Using a Novel A-Star Algorithm And Neuro Fuzzy Volosencu, C. and Curiac, D.(2013), Efficiency improvement in multi-
Approach, International Journal on Current Issues of Computer sensor wireless network based estimation algorithms for distributed
Science, Vol. 1,No 1 pp 21-27 . parameter systems with application At the heat transfer, EURASIP
Merlin, C., Feng,C. and Heinzelman, W.,(2009) Information-Sharing Journal on Advances in Signal Processing , springer Vol 1,no.4,pp
Protocol Architectures for Sensor Networks: the State of the Art and a 1-17.
New Solution, Int. Journal of Mobile Computing and Communications Wang, J., Yang, X., Ma, T., Wu, M., and Kim J (2012), An Energy-
Review Vol.3,No.4,pp.26-38, efficient Competitive Clustering Algorithm for Wireless Sensor
Mhemed, R., Aslam, N., Phillips,W. and Comeau ,F. (2012), An Energy Networks using Mobile Sink, In the Proceeding of International
Efficient Fuzzy Logic Cluster Formation Protocol in Wireless Sensor Journal of Grid and Distributed Computing, Vol .5, No 4.
Networks, The 3rd International Conference on Ambient Systems, Wang, N. and Zhu, H. (2012),An Energy Efficient Algorithm Based on
Networks and Technologies(ANT), 27-29, , Niagara Falls, Ontario, LEACH Protocol, International Conference on Computer Science and
Canada, August 2012.Northumbria Research Link: Electronics Engineering, pp 339-342.
http://nrl.northumbria.ac.uk/7042/ Yan,W., Zou, Z.and Wang,H.(2010), Adaptive neuro fuzzy inference
Mishra S., Jena L., Chakrabarty, A. and Choudhury , J.,(2012), Fault system for classification of water quality status, International Journal
tolerant Multi Cluster head Data Aggregation Protocol in WSN of Environmental Sciences ,Vol 22, No 12,pp.1891-1896.
(FMCDA),In the proceeding of International Journal of Young ,Z and Pei ,Q.(2012), A Energy-Efficient Clustering Routing
Technological Exploration and Learning (IJTEL),Vol.1,pp 32-36. Algorithm Based on Distance and Residual Energy for Wireless
Mon ,Y.J, Lin ,C.M and Rudras ,I.J.(2013) , ANFIS-based Wireless Sensor Networks, In the proceedings of International Workshop on
Sensor Network (WSN) Applications for Air Conditioner Control, Int. Information and Electronics Engineering (IWIEE),Vol 29,pp 1882-
Journal of applied sciences Acta Polytechnica Hungarica Vol. 10, 1888.
No. 3, pp.5-16. Ziasabounchi ,N. and Askerzade, I.(2014), ANFIS Based Classification
Model for Heart Disease Prediction, International Journal of
Electrical & Computer Sciences IJECS, Vol.14, No 02,pp 7-12.

3875 | International Journal of Current Engineering and Technology, Vol.4, No.6 (Dec 2014)